[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: making psamp export congestion-aware
>Unreliable transport: The export of measurement records does not
>require reliable export.
I too feel unreliable export is enough.Netflow is an well experimented
one using unreliable support.But,How netflow floks account for reliability
for some applications like billing & accounting .
>In fact, retransmission of lost measurement
>packets could consume additional network resources and would require
>state on the device generating the records. The device would have to
>be addressable, able to receive and process acknowledgments, and to
>store unacknowledged data. These requirements would be a significant
>impediment to having uniform support for measurement on high-speed
>interface cards in IP routers. Instead, we propose that PSAMP devices
>can support a unreliable export mechanism.Sequence numbers on the
>packets or the measurement records within the packet would indicate
>when loss has occurred, and the analysis of the measurement data can
>account for this loss. In some sense, packet loss becomes just
>another form of sampling (albeit a less desirable, and less
>controlled, form of sampling).
To bolster Jen's points of using a unreliable transport ..we can also
look at the granularity of the data being exported.I think reliable
export can be used when one sent's aggregated data(as the extra work
done to make it reliable doesnt make a big difference) which is not the
case with psamp.My another important point is:Exporting will itself eat
lot of router processing and so additional burden of keeping the state
will make the processing worse.
Also,some of the discussion in IPFIX will give us more idea into this aspect.
Is their any difference between IPFIX Vs PSAMP with respect to congestion-aware
export.The reason i am raising this issue is ..there is a strong lobby out there
to support a reliable export.
How do you account for failure of collector(another strongly debated topic)
We will surely need some extra mechanisms or use SCTP(well sugessted one but
less implemented )
- Senthil
--
to unsubscribe send a message to psamp-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/psamp/>
h{.n+zwZ,jު笶l_0m(ۧrz)ڲ)b欶z^w&rzmȞ+b?Ʀ